Needling sensation: explanation of incongruent conclusion drawn from acupuncture FMRI study.

Dear Editor: Whether or not acupuncture intervention involves neurologic effects has recently been explored via functional magnetic resonance imaging (fMRI) experiments commonly consisting of multiple experimental blocks with ONand OFF-condition alternations. Such fMRI experiments normally employed a block interval of more than a minute.1–3 However, the psychophysical response to acupuncture, a needling (or de qi) sensation (a sensation including sourness, numbness, heaviness, and distension surrounding the area of needle insertion), introduced by the acupuncture stimulus conducted in one current trial can be carried over to the next trial. This carryover effect might alter the baseline condition of the following trials and thus affect data analyses, possibly resulting in incongruent conclusions. Here, we demonstrate the carryover effect with a psychophysical test using a single block of acupuncture needling.4 Subjective psychophysical responses were obtained from 16 subjects (6 females and 10 males, 27 2.7 years old). The subjects received acupuncture at the acupoint of left ST42 (Chong Yang). In this experiment, the needle was inserted and twisted clockwise and counterclockwise at a frequency of 1–2 Hz to produce a needling sensation. The subjects were requested to continually report this needling sensation verbally every 10 seconds, for up to 1 hour. The needling sensation was scored from 0 (no de qi at all) to 10 (the strongest de qi sensation a subject can endure). The result showed that the initial de qi was the strongest reaction, which then reached its plateau at 20 seconds and started to drop at 2 minutes (Fig. 1). This strongly suggests that if the block duration is less than 2 minutes in a multiblock fMRI experiment, the evoked brain responses registered to the current acupuncture stimulus would be prolonged enough to perturb the baseline of the subsequent trial(s). This can further affect the ON/OFF contrasting in the statistical analysis. In other words, a multiblock design with its OFF-condition lasting longer than 2 minutes for each block or a single-block design will be more suitable for conducting an fMRI acupuncture study. In the future, we will demonstrate empirically the influence of carryover effects on an fMRI study involving acupuncture stimulation with a larger sample size and compare the result from multipleand single-block experimental designs.